Optimizing performance analysis for synchronous dataflow graphs with shared resources

Thiele,; Ernst, F.

doi:10.1109/date.2012.6176549

Cited by 3 publications

(8 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, we assume a flavor of data flow called Single-Rate Data Flow (SRDF) [12] in which every actor firing (activation) consumes and produces a single token on each of its incoming and outgoing edges respectively. Note that our technique can be extended to accommodate any synchronous / cyclo-static data flow graphs (see section V-3 or refer [16]). Since tasks in a streaming application are event triggered and activate as soon as they receive the required input, we consider a selftimed execution of our data flow actors so that they fire as soon as they have a token on all incoming edges.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Analyzing preemptive fixed priority scheduling of data flow graphs

Lele

Morena

Bastos

et al. 2014

2014 IEEE 12th Symposium on Embedded Systems for Real-Time Multimedia (ESTIMedia)

View full text Add to dashboard Cite

Data flow graphs can conveniently model embedded streaming applications (ESAs) that are typically implemented as networks of concurrent tasks having an iterative pipelined execution, where the activation of each task may be conditioned by intra-and inter-iteration data dependencies. We propose a novel analysis approach for preemptive Fixed Priority Scheduling (FPS) of multiple ESAs assuming a fixed mapping of tasks onto the processors of the underlying Heterogeneous Multi-Processor System-on-Chip (HMPSoC).The tasks of an ESA are event activated, have varying execution times, and participate in cyclic dependency chains such that they may not have an activation pattern that can be depicted using traditional periodic / sporadic event models. Instead we propose to characterize the data flow graphs of ESAs to upper bound the load they impose on a processor and use it to compute the worst-case response time of an actor executing on that processor at a lower priority. We show that ours is a generic approach for analyzing FPS of data flow graphs. We also propose a refinement of our technique for graphs with a dominant periodic source. We demonstrate our improvement over the state-of-theart FPS analysis for data flow in our experiments.

show abstract

Section: Introductionmentioning

confidence: 99%

“…Contribution: Existing FPS analysis techniques [8], [9], [14], [16] attempt to extract a periodic-with-jitter based event model for graphs. Our work characterizes the graphs to induce the worst-case effect FPS into the graphs to perform our analysis.…”

Section: Introductionmentioning

confidence: 99%

Analyzing preemptive fixed priority scheduling of data flow graphs

Lele

Morena

Bastos

et al. 2014

2014 IEEE 12th Symposium on Embedded Systems for Real-Time Multimedia (ESTIMedia)

View full text Add to dashboard Cite

show abstract

“…Our approach differs from [17], since modeling schedules directly into SDFGs enables us to use dedicated analysis for dataflow graphs. Moreover, the technique of [17] can only handle an SDFG with limited types of cycles, such as cycles formed by self-edges or the back edges modeling the buffer capacity between two actors. However, staying in dataflow domain, as is done in our technique, does not impose such a limitation on the graph structure.…”

Section: Related Workmentioning

confidence: 99%

“…For shared resource analysis, they use event-models [18] which is based on realtime-calculus [19]. Our approach differs from [17], since modeling schedules directly into SDFGs enables us to use dedicated analysis for dataflow graphs. Moreover, the technique of [17] can only handle an SDFG with limited types of cycles, such as cycles formed by self-edges or the back edges modeling the buffer capacity between two actors.…”

Section: Related Workmentioning

confidence: 99%

“…However, for non-preemptive schedules, our results are tighter and our techniques require less analysis time. The authors of [17] have shown that an SDFG can be used to consider an application with resource sharing possibilities; they perform buffer sizing under a throughput constraint considering a given schedule for the actors using a shared resource. For shared resource analysis, they use event-models [18] which is based on realtime-calculus [19].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Schedule-Extended Synchronous Dataflow Graphs

Damavandpeyma

Stuijk

Basten

et al. 2013

IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst.

View full text Add to dashboard Cite

Abstract-Synchronous dataflow graphs (SDFGs) are used extensively to model streaming applications. An SDFG can be extended with scheduling decisions, allowing SDFG analysis to obtain properties like throughput or buffer sizes for the scheduled graphs. Analysis times depend strongly on the size of the SDFG. SDFGs can be statically scheduled using static-order schedules. The only generally applicable technique to model a static-order schedule in an SDFG is to convert it to a homogeneous SDFG (HSDFG). This may lead to an exponential increase in the size of the graph and to sub-optimal analysis results (e.g., for buffer sizes in multi-processors). We present techniques to model two types of static-order schedules, i.e., periodic schedules and periodic single appearance schedules, directly in an SDFG. Experiments show that both techniques produce more compact graphs compared to the technique that relies on a conversion to an HSDFG. This results in reduced analysis times for performance properties and tighter resource requirements.

show abstract